Advantages of Fixed Bias

[Michael kiley]

ADVANTAGES OF USING FIXED BIAS IN VACUUM TUBE AUDIO POWER AMPLIFIERS
by Michael Kiley

Using fixed, as opposed to self, or cathode resistor, bias in a vacuum tube audio power amplifier has the following advantages:

1. As the name says, the bias point for the output tubes is fixed and does not change with cathode current or signal level.

2. The voltage dropped across a large cathode resistor, multiplied by the cathode curent, which you would have with self bias, is not wasted as heat. Also, because the cathode is at or very near ground potential, almost all of the plate supply (B+) voltage is directly across the tube itself.

3. With fixed bias, the idling current is usually lower than it is with self bias. Therefore, the tubes run cooler and the cathode emissive coating is not depleated as rapidly.

4. An electrolytic capacitor (cathode bypass capacitor) is eliminated from the signal path. Such capacitors may be detrimental to sound fidelity.

Fixed bias will give more power output, less distortion, more efficient operation, and longer tube life. For these reasons, almost all public address and other commercial duty tubed amplifiers were fixed biased. However, in addition to adding a negative voltage source, other considerations must be taken into account when building a fixed bias amplifier or converting an existing self biased amplifer to one with fixed bias.

1. A means must be provided to measure the idle current through each tube and adjust each tubes's bias voltage so the current is at its proper value and both (or all) tubes are drawing nearly equal current. Even matched sets of tubes may not be matched at all points on their grid voltage vs. cathode current curves, and these may drift apart over the tubes' life. Generally, a low value (10 to 47 ohms) resistor is inserted between each tube's cathode and ground, and a test point is connected to the cathode end of that resistor, making this easy to monitor with any voltmeter. The normal drop across this resistor is only a volt or two, at most. To adjust the bias voltage, a trimmer potentiometer is provided to vary it about +/- 30% of its nominal value. With two tubes in push-pull, one control may vary the bias on both tubes while another pot may adjust the balance between the two tubes. The balance is adjusted for zero voltage difference between the cathodes, while the bias is used to set the voltage on either cathode to equal that of the proper current. The bias should be adjusted any time tubes are replaced and once or twice a year in between.

2. Grid emission is more critical with fixed bias. Therefore, the value of the grid resistor, plus the resistance of the bias supply and adjustment circuit should not exceed 100K ohms. 6BQ5/EL84's and 6CA7/EL34's are rated for 300K, but given the sometimes poor quailty control in making tubes these days, even these types should be kept to 100K. This means a medium mu driver will be needed, with its plate (and cathode, if the driver is a split-load phase splitter) load resistors at 47K or less. If converting a Mullard 520 to fixed bias, the 12AX7/ECC83 driver will have to be changed to an 12AU7/ECC82 or a 6CG7, and the cathode, plate, and B+ dropping resistors lowered. Look at the long-tailed phase splitter stages used by Macintosh, Marantz, and Eico's higher power amps. for possible values. The coupling capacitors to the grids of the output tubes should be at least .1uf (100nf) or perhaps .22uf (220nf).

3. To prevent output tube meltdown if the bias fails, fuse the B+ line with a slow blow fuse, sized at about 150% of the full output plate + screen grid (G2) current. 250ma. is ideal for a pair of 6L6GC, 5881, or 6CA7/EL34 types. Be sure the screen grids are powered on the load side of this fuse so they will also be disconnected if the fuse blows.

4. Idling to full power output current swing from the B+ supply will be greater with fixed bias, so supply regulation is more critical. Use larger filter caps. and a low resistance rectifier.

Self bias, in addition to its simplicity and low cost, is largely self adjusting if each tube has its own cathode resistor and bypass capacitor, as the Mullard 520 did. Also, self biased output stages may go into clipping more gradually as the voltage drop across the resistor increases at full output, and may recover faster from overdrive. This may be useful in some guitar amplifier applications.

In general, fixed bias is better for amplifiers producing over 20 watts of power.

MJK 8/28/11

[Geek]

Hi,

Can be true for pentodes, but fixed bias in a near-vertical curved triode (45, 2A3, 300B) is looking for trouble

Cheers!

[dcgillespie]

Also -- In a true class A design (PP or SE), the type of bias used falls out of the equation, aside from the B+ lost across the cathode resistor.

Dave

[Michael kiley]

I have seen 2A3 listed in the tube manual as giving 15 watts (2 in PP) with fixed bias, and only 10 W with self bias. I have only worked with tetrodes and pentodes, but I have seen triodes in fixed bias. If the stage is running true class A, where the current is hopefully unchanging, this may be moot. Self bias is more self-adjusting, as I mentioned at the end of my article, so some unusually finicky tubes may be better with it. But the more common tubes will give more performance with fixed bias. Dave Manley made an even stronger stand on this in his Vacuum Tube Logic Book

Mike